MEMORANDUM REPORT NO. 1840

OPTICAL METHOD FOR ANALYSIS OF

ATMOSPHERIC EFFECTS ON LASER BEAMS

by

Paul H. Deltz

July 1967

Distrihutior of this document is unl:'ited.

U. S. ARMY MATERIEL COMMAND _-

BALLISTIC RESEARCH LABORATORIESABERDEEN PROVING GROUND, MARYLAND

E - I N

U.S. AF4Y BALLISTIC RESEARCH LABORATORIESABERDM PROVING GROUND, MARYLAND

6 April 1967

ERRATA SUET

for

BRL Memorandum Report No. 1840 entitled "Optical Method for Analysis ofAtmospheric Effects on Laser Beams", dated July 1967.

1. Page 16, line 9. Change 5.0 to 0.50

2. Page 18, change the following entries in Table II to read:

%C (calculated)

Afternoon 0.38 2.0 x l0"' m"/3

F 5x1 -1/3Night O.lL 7.5 x 10 -

3. Appendix Re Figures A-I to A-4, A-14: The photographs of the beamcross section should be oriented to show the step wedge along the

-left margin rather than the bottom margin. See text (pages 12-13)for optical system description.

BALLISTIC RESEARCH LABORATORIES

MEMORANDUM REPORT NO. 1840

JULY 1967

OPTICAL M1HOD FOR ANALYSIS OFATMOSPHERIC EFFECTS ON LASER BEAMS

Paul H. Deitz

Ballistic Measurements Laboratory

This material was presenttJ at the Symposium on"Modern Optics," sponsored by Brooklyn PolytechnicInstitute and held at Waldorf-Astoria Hotel, New YorkCity, 21-294 March, 1967.

Distribution o' this document is unlimited.

IFNE5M DT&E Project No. IP523801A286

ABERDEEN PROVING GROUND, MARYLAND

-¶7

BALLISTIC RE3 EARCH LABORATORIES -

WEMORANDUM REPORT NO. i84o

PHDeitz/sJw

Aberdeen Proving Ground, Md.July 1967

OPTICAL METHOD FOR ANALYSIS OFATMOSPHERIC EFFECTS ON LASER BEAMS

ABSTRACT

This report describes the design of an inatrumentation system for

the study of the effects of atmospheric turbulence on a collimated laser

beam under near-earth conditions.

The instrumentation consists of a helium-neon laser with optical

collimator and a receiving system of 24-inch aperture with narrow; band-

pass filter.

The Appendix presents examples of beam crosw section patterns for

different propagation conditions. The method of analyzing spatial

intensity distributions of these patterns is described.

3

JI

TABLE OF CONTENTS

Page

ABSTRACT . . . . . . .. . 3

LIST 0 IOUEES . ...... . . . . . . . . . . 7

INTRODUCTION ........................... 9

INSTRtUNTMATION . . . . . . . . . . . . . . . . . . . . . . . . . 10

CALIBRATION 1

ANALYTIC )THOD .... . . . . . . . . . . . . . ..... .

THEORY AND RESULTS . . . . . . . 13

SUWARY AND CONCLUSION ....... . . . . . . . . . . .18

RE XCS .. . . . .. .. .. .. .. .. .. ..... 20

APPENDIX . . . . . . . . . . .... 21

DISTRIBUTION LIST .. . . . . ..... . . . . . . . . . . .47

'5

woI

LIST OF FIGURES

1 Ray diagram of )ptical receiver ............ 9

2 Electromagnetic propagation range with 30-foot sensorstations . . . . . . . . . . . . .. . .. . . . ... 12

A-1 Beam cross-section photograph taken under highscintillation conditions ............ ... 23

A-2 Beam cross-section photograph taken under mediumscintillation conditions .. ........ .... 2. 4

A-3 Beam cross-section shoving unusually large cellstructure . . . . . . . . . . . . . . . . . . . . . . . 25

A-4 Sample of beam cross-section data used for quantitativeanalysis (afternoon propagation) . . . . . . . . . .. 26

A-5 A computer-derived characteristic curve . . . . . . . .. 27

A-6 Horizontal intensity prdfile of beam cross-section . . . 28

A-7 Temporal frequency times spectral density vs temporalfrequency . # . . . * .. ... . .. . . . .. . . .. 29

A-8 Temporal frequency times spectral density vs temporalfrequency normalized to a normal wind component of1 meter/second . . . . . . . . . . . . . . . . . . . . 30

A-9 Fourier power spectral density plot of horizontalintensity profile (afternoon propagation) . . . . . . . 31

A-10 Spatial frequency times power density spectrum vsspatial frequency (afternoon propagation) . . . . . .. 32

A-11 Spatial frequency times smoothed power density spectrumvs spatial frequency (afternoon propagation) ..... 33

A-12 Spatial frequency times averaged horizon~al and verticalpower density spectra vs spatial frequency (afternoonpropagation) . . . . . . . . . . . . . . . . . . . .*. 34

A-13 Spatial frequency times averaged smoothed horizontal andvertical power density spectra vs spatial frequency(afternoon propagation) . . . ... . . . . . . . . ... . 35

71

LisT oFnFIuRES (Continued)

Page

-A-14 Sample of bean cross-section data used for quantitativeanalysis (night propagation, .............. 36

A-15 Fourier power spectral density plot of horizontalintensity profile (night propagation) . . . . . . . . . 3T

A-16 Spatial frequency times power density spectrum vs spatialfrequency (night propagation) .............. 38

A-17 Spatial frequency times smoothed power density spectrumvs spatial frequency (night propagation) . . . . . . . . 39

A-18 -Spatial frequency times averaged horizontal and verticalpower density spectra vs spatial frequency (nightpropagation) . . . . . . . . . . . . . . . . . . . . . .

A-19 Spatial frequency times averaged smoothed horizontal andvertical power density spectra vs spatial frequency(night propagation) .................. 4.1

A-20 Spatial frequency times smoothed power density spectrumvs spatial frequency for a series of increasing one-dimensional apertures f'rom afternoon horizontalintensity scan. . . . . . . . . . . . . . . . . . . . . 42

A-21 Distribution curve for afternoon horizontal intensityscan .. . . . . . .. .. .. .. ... . 43

A-2P Distribution curve for night hor'zontal intensity scan . . 44

A-23 Cumulative distribution curve for afternoon horizontalintensity scan . . . . . . . . . . . . . . . . . . . . 45

A.-24 Cumulative distribution curve for night horizontalintensity scan . . . . . . . . . . . . . . . . . . . 46

8

9

INTRODUCTION

Nearly all optical systems involve the transmission of electro-

magnetic radiation through an atmosphere. For most systems operating in

controlled atmospheres, the medium is assumed to be honogeu:'ous and its

effect on system performance negligible. However, for long paths

through the &tmosphere the medium is not homogeneous and the index of

refraction varies along the beam; this variation can be characterized

statistically. Because of the effect of the medium on the transmitted

beams, the performance of many optical systems is finally limited by

atmospheric conditions.

To study these effects, the Ballistic Research Laboratories (BRL)

have designed and built a high resclution, high stability optical

receiving system. It is designed specifically for the study of light

propagation in the near-esrth atmosphere. A schematic drawing of the

system is shown in Figure 1. This report describes the design of the

system and the method ueed to analyze atmospheric effects on a collimated

light beam recorded by the system. Results of the analy3ia show that the

effects correlate well with those predicted by Tatarski. This system

should be a valuable aid to a better understanding of atmospheric effects

on optical systems.

0PTrCL.

FmLAT 4 I*AI, M A""U

FinsI min

is.~~~ 11 14WEM

S~Figure 1. Ray diagram of optic•.] recelvel

*

9

PL-A

INBTEU3INDATION

-Th light source is a helium-neon laser with a power of 10-aulliwatts.

T eoput-can be attenuated by a polaroid filter and monitored at the

Slaser. -Th attenuated beam is sent into a collimator with a spatial

-filter located at the focus of Zhe eyepiece to remove intensity variations.acrossthe lUer beam. The £-inch objective lens is defocused slightly

6s that the transmitted Airy disc is somewhat larger than the receiver

"'Aprture at the range of 549 meters. The laser and related optics are

held on an aimable mount. This system is located in a small trailer,

isodly ground supported.

-• h optical system is contained in a 25-foot trailer making possible

easy transportation from one site to another. On location, the heavily

reinforced van :og jacked free from its wheels. Inside the trailer, the

optical system is mounted on an 8-inch I-beam frame. This frame is

lifted free of the trailer floor by means of a three-point jacking system

supported directly on the ground.

Collimated light impinging or a parabolic mirror 2 feet in diameter

with -a 10-foot focal length is reflected off-axis toward a plane mirror

and then a focus. Arz aperture is placed a& this focus which restricts

the acceptance angle of the system to about t-wo tenths of a degree.

Niext, an f/4, E-inch focal length lens is placed one focal length from

the aperture. Thus, the emerging light is collimated, making possible

introduction of an interference filter which eliminates neerly all of any

kemaining background illumination. Beyond this first lens and filter is

an image of the spatial intensity distribution at the 2-foot p&rabolic

mirror. Alongside this image, an illuminated step wedge is projected in

conjunction with fiducial lights to enable frame-to-frame correlation

between a timed series of photographs. Both the intensity distributions

of the parabolic mirror and the step wedge then are imaged by a second

lens to a sensor plane at a magnification compatible with a chosen

format.

10

-I

___________~-•--, .--.

Sii

CALIBRATION

An electromagnetic wave is completely characterized by its amplitude

and phase. The optical receiver was set up to record the amplitude of

the impinging radiation in terms of its related intensity. A 35-milli-

meter motion picture camera was placed in the film plane of the receiver

to record the spatial intensity pattern of the incident radiation along

with the calibrated step wedge. The wedge was calibrated at the film

plane to an accuracy of 5 percent. The intensity variation across the

extent of the entrance aperture of the receiver was measured by illuml-

nating the parabolic mirror with the far-field pattern and scanning the

image plane. Over the central 18 inches of the mirror, the intensity

variation was less than 6 percent. A geometrical distortion and resolu-

tion test .as made by placing a test screen in front of the receiver and

back-illuminating it with the divergent laser beam. A grid array of

holes on 10-centimeter centers was drilled in the board of quch size as

to give a 2 millimeter diameter Fraunhafer diffraction pet,*- on the

parabolic mirror surface. Pictures were takeu )f this pattern and Imeasured to a 1-micron precision on a comparator, Accurate measurements Al

were made of the grid array, and a least squares fit computer program was

run on these data. The standard deviation of the porition of the image

points related to object space and compared with the grid array was 0.7

millimeter with a resolution of 0.25 millimeter or better for the total

receiver. A recently modified fiducial mark projection system gives

positional accuracy in the image plane to withain 10 microns. In object

space, this corresponds to a positional accuracy of 1.3 millimeters.

ANALYTIC METHOD

The type of optical data being sought initially was beam intensity

as a function of position ov a horizontal and a vertical diameter of the

parabolic mirror. Varlous films had been tested to determine the

dynam•c range and gama of the linear portion of the characteristic curve

of the film. For each type of film and exposure time, a particular

absolute intensity variation of the step wedge covers the dynamic range

ll

or usable portion of the film curve. One of the brighter steps gives an

intensity at the top of the dynamic range of the film. Neutral density

filters were inserted at the optical receiver until the most intense

radiation correspOnds to the intensity of the brighatest usable step.

,After the data were taken and the film developed, the individual frames

were scanned consecutively on a densitometer, IuLe scans were always over

the same 2 diameters of the parabolic mirror. The densiti3s as a function

of positicn were recorded and then converted to relative intensities

through the characteristic curve of the film obtained from the step

vadge.

A series of trial photographs were made to determine the feasibility

of the technique. The path length is 549 meters, averaging about 2 meters

above short grass. The location was the Electromagnetic Propag&tion Range

at the BRL. The range featares an extensive array of meteorological

sensoirs with an electronic data acquisition system (Figure 2).

Figure 2. Electromagnetic propagation range with30-foot sensor stations

Figure A-i in the Appendix shows a mottled pattern typical for a

day with high scintillation. The step wedge extends vertically in the

optical system. Since tbe system incorporates an optical flat, there

is a horizontal, but not a vertical, image inversion. Thus, the image

can be understood to be the intensity distribution that would be seen if

12

a 2-foot section of ground glass were to be substituted for the parabolic

mirror with the observer located behind the ground glass looking back

toward the source. This photograph was taken in the sumner during a

mid-afternoon. The exposare time for all of the photographs is 1/500

second. There was, therefore, no image blur in the horizontal direction.

Figure A-2 is a photograph taken at night under medium scintillation

conditions. Since the atmosphere was fairly quiet, the high degree of

ccherence of the laser was maintained, resulting in the characteristic

diffraction effects from dust particles on the lenses.

Figure A-3 shows an example of a fleeting optical effect observed

after dusk one summer evening. The pattern consisted of large, slowly

changing cells which built up gradually, lasting in this conditi-An for

about a minute. The pattern broke up quickly into much higher spatial

frequencies; once more built up to the large cell structure for a

minute; and finally broke up altogether.

Figure A-4 shows one of two photographs of which a quantitative

analysis has been made. The step wedge at the left of the photograph was

scanned and a compueer-fitted characteristic curve derived (Figure A-5).

The relative densities are plotted vertically and the logarithm of the

exposure is shown to the right. The antilogs are also given The

extremes of the exposure are indicated for the horizontal scan through

the mirror. Using this curve, the oatput of the densitometer was

computer-corrected to relative intensities, and plotted (Figiure A-6).

THEORY AND) RESUM~S

Tatarski1 predicts the two-,imensimal spatial spectral densities

in a plane perpendicular to the direction of propagation for a plane,

monochromatic wave. The spatial spectral densities are postulated to be

isotropic and to be a function only of the wavelength, A, and the path-

length, L. For the case where the wind diraction angle relative to the

optical path is much larger than the quantity (A/L) 1 /2, the intensity

structure is assumed to be "frozen-in" and carried ar.oss the optical

13

path-by the normal wind component, vn. Tatarski thus translates from a

twod siO spatial, spectral density domain to a one-dimensional

-InW Tatsrski's aeriment, a weakly divergent incandescent beam ispropagated oer�a. hcmogeneous path within 2 meters of the ground. Thereciv.• �i�i•gtit i•snitored by a photocell with an aperture small enough

- (" mill tes) to resolve the highest spatial frequencies. As the normal-wid c-mponent takes- the turbuLmt medium across the beam path, a one-

x- nioinal-intensity profile is recorded. Since the optical path is"nee:•n.r thieo , the vertical wind component is necessarily small, so'hat the profile is predominantly horizontal. The magnitude of v ismoitored during the experiment. Figure A-7 shows the results of suchmeasurement. The output of the photocell is sent into a time spectrumanalyzer. The power, W(f), is multiplied by temporal frequency andt plotted versus the log2 of the temporal frequency. As can be expected,-asn goes to higher values, the peak power point of the curve also moves-to higher frequencies. Next, all these curves, along with data taken at-different patblenths, are normalized to a vn of 1 meter per second anda path of 1,000 meters (Figure A-8). The curves then assume, generally,the same shape.

Tatarski's time spectrum is related to the space spectrum by

TinS *vni n

where the space spectrum, S, is measured in cycles per meter. When thespace spectrum is thus measured, it is equivalent to the time spectruminherently normalized to ev of 1 meter per second. To test thisnreletionship, a power-Olmsity, spectrv analysis Vas made using thehorizontal intensity profile shovw in Figure A-6. Figure A-9, then, shows

tht relative powers as a function of spatial frequencies. Next, the powerVa multiplied by frequency and plotted on a logarithmic abscissa

(Figure A-10). Since the spectra have frequency plotted in cycles per

meter and are thus normalized for a normal wind component of 1 meter per

S14

|

second, the space spectra units convert directly to time spectra units.

The roughness of the curve Is indicative of the shortness of the sample

length. To get a more definitive plot, many such samples should be

analyzed and averaged. In order to approximate the effect of a longer

sample length, a smoothing program was applied to Figure A-10 resulting

in the curve shown in Figure A-11. This reveals a much greater likeness

to the Tatarski time plots but the spatial spectra were normalized by

different schemes and the ordinates cannot be compared. With this method

of data acquisition there is no time lag from the beginning of the

Sintensity scan to the end so no assumption about the degree to which the

turbulent cell structure is "frozen-in" is necessary to infer the nature

k of the true spatial frequency characteristics. Under the assumption that

the photograph was indeed isotropic, The unsmoothed vertical and hori-

zontsl power density spectra were averaged, multiplied by frequency, and

plotted on a logarithmic abscissa (Figure A-12). These combined spectra

units were then Etmoothed (Figure A-13), and it was this curve that was

used for comparison to the theory.

By the same process, a photograph taken during a night propagation

run was also analyzed (Figure A- 1 4). As in the earlier photograph taken

at night, Figure A-2, the effect of highly coherent light operating on

the optical receiver can be z een as diffraction rings in the image plane.

Figure A-15 shows the power density spectrum, and Figures A-16 to A-19

show the frequency plots smoothed and averaged as before.

Tatarski predicts that the frequency at which the maximum power

occurs is given by the formula

f = A n (2)m (XL)

where f is defined as te average of the two frequencies occurring atM

one-half the peak power. Theoretically, A is given as 0.55, while

experimentally Tatarski obtains a value of 0.32. Using Figures A-13 and

A-19, an average A of 0.222 was obtained for the two photographs.

15

Although not explicitly stated later in Tatarski, f is derived for am

pliane wave source. The difference between the theoretical and experi-

mental values in not explained. If, Ln a very simple model, all of

the diffraction is assumed to take place at a plane 'located midway between

-the tranmmitter and receiver, then the spatial frequencies will be scaled

down at the receiver by the ratio of the beam width at the path midpoint

to the beam width at the path terminus. As transmittc4 light is changed

from a collimated beam to a point source, B goes from a value of 1.0 to

5.0. Table I shows this "divergence factor" for both experiments. This

then is to be applied as a correction factor to the coefficient 0.55 for

a plane wave. The product of these two is in the column titled "A,

calculated." This value can be compared with the measured value of A.

Table I

width of beam at midpointSA B(0.55) B =idth of beam at terminus divergence factor

Bcalculated Acalculated measured

Plane Wave 1.0 0.55 --

Tatarski o.6 0.33 0.32

BRL 0.50o4 0.277 0.222

The importance of resolving the highest spatial frequencies in

object space was inferred earlier. The influence of the diameter of the

photocell aperture on the frequency distribution has been predicted by

Tatarski and measured by others. 2 With the present demagnification from

object space to the image plane and the size of the scanning aperture ofthe densitometer, the effective aperture size in object space is 1.26

millimeters. To show the effect of an increasing one-dimensional aper-

ture, the intensities shown in Figure A-6 wvre averaged over windows

with widths of 5, 9, 17, and 33 samples. Each of the smoothed scans wassubjected to frequency analysis, smoothed, and plotted (Figure A-20).

L The smoothed frequency times power plot of Figure A-11 is also shown for

16

comparison. The effective sizes of the apertures in increasing widths

are 1.26, 5.16, 9.0?, 16.9, and 32.5 millimeters. As the widths are

increased, both the decrease and shift in peak power and the loss of

high frequency response are readily apparent.

The shape and maxima of the spatial frequency spectra are not

supposed to vary for a given pathlength or wavelength. However, the

intensities, which are postulated to be log normally distributed, are

predicted to exhibit a variance proportional to the fluctuations in the

index field. Tatarski's formula giving this relationship isi 2 kT/6 1. (/6

0 1.23 Cn ()

where a is the standard deviation of the log normal distribution, C is

the index structure constant, k is the wave number of the light, and L

is the pathlength. To test this relationship, the horizontal intensities

from Figures A-4 and A-I4 were each normalized by division of their

average intensity. The percent number of occur-.ences was then plotted

versus the normalized intensities to form distribution curves (Figures

A-21 and A-22). Figure A-21 from the afternoon propagation data shows

a much greater skewness, indicating higher scintillation conditions.

Next, the integral of each of these curves was taken to yield theF cumulative distribution curves. The integrated percent occurrences were

plotted on probability paper versus the log of the normalized intensities

(Figures A-23 and A-24). Both sets of data conform closely to log normal

distributions. The divergence at the extremes of the curve in Figure A-23

can be interpreted in the following way. The first point was taken from

the very edge of the mirror where the calibration has indicated a sharp

attenuation of intensities due to vignetting. The intensities at the

upper end of the curve came from the top of the characteristic curve.

Since they were forcing the dynamic range of the film, their true values

were likely higher than those indicated. Although no Cn was measured

for the two cases, the cumulative distribution curves were used to derive

the standard deviation. Equation (3) was sojved for Cn, and the appropri-

ate values were inserted; Table II lists the results.17

li A

Tab!, II

C (calculated)0 n

Afternoon 0.852 1.96 x 1O-7 m71/3

Night 0.332 7.63 x 10-8 m1/3

3Weak turbulence c = 8 x lo-9 071/3n

Intermediate turbulence C = 4 x lo-8 m71/3n

Strong turbulence Cn = 5 x 10-7 m71/3

Comparing the calculated values of C with those given br Davis3 fornthree representative cases indicates reasonable agreement for the inter-

mediate to strong scintillation conditions.

SUMMARY AND CONCLUSION

Initial data analysis of the spatial intensity distributions from

the optical receiver shows a h_ gh degree of correlation to the theoreti-

cal predictions of Tatarski. When the divergence of the transmitted beam

is taken into consideration, the predicted frequency of maximum power

compares favorably with the measured value. Widening of the sampling

aperture is shown both to attenuate the high frequency resolution and to

shift the peak power point to lower frequencies. Calculated values of

the index-structure constant from the measured variance of the intensity

profiles conform to the proper orders of scintillation magnitude.

During normal data runs, a time series of photographs is taken to

make the results statistically meaningful. Since in this particular

approach the scanning is not a function of the normal wind component,

any intensity profile can be examined, including the vertical, allowing

a check of the assumption of optical isotropy both by electronic reduction

and direct optical correlation techniques. By this method, the spatial

frequency domain may be converted to a temporal frequency dome .n which

is inherently normalized for a wind component normal to the direction of

'L8

beam propagation of 1 meter per second. No assumption about the degreeto which the turbulent cell structure is "frozen-in" is necessary toinfer the nature of the spatial frequency statistics. In addition, thehigh resolution and good signal-to-noise characteristics of the receivermake accurate measurements of beam pointing reliability possible forvarious system applications.

This BRL approach will help to give a better quantitative under-standing of the constraints placed on nearly all optical systems workingin the atmosphere.

ACKNOWLEDGMENTS

The author gratefully acknowledges the assistance ofR. B. Patton, Jr., for his contributions to the data analysis and toDonald Portman and, through him, Robert Hufnagel, for their thoughtsconcerning the beam divergence concept.

1

19

-x

. -V.T.. Tatar"kl, "Uve PzaMatLon in a Turbulent Nuditm,"

2.• Ss D. 4. Portman, F. C. 31l.i, 1. Pmnwr and V. I. Roble, "Journal-ot 6.' a Itsaea'eh," Vol, 67,T3M, 1962.

3. IA. hU uAIlq~ed OptIcas Vol. 5, No. 1 Januaz 1966, p -uAl.

20

iI

* I

• • 20

.APPENIDIX

iIFIGURE

(Atmospheric E ;eets on Laser Beams)

21

FiueA1 emcosscin htgahtknudrhg

scinillaion ondiion23i

FgrA-2. Bean crs-seto phtgrp taken under mediY~q.um

scintllatin conition

-24

N?

Ii

I

I

Figure A-2. Bean. cross-section photograph taken under mediumsci ntill ]ation condi tions

24

; *1

7' 3t _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __. I

Figure A-3. Beam cross-section showing unusually lrge cell structure

25I 7I

Figurt A-4. Sample of beam cross-section data used for quantitativeanalysis (afternoon propagation)

26

08

tJ4

10 0

8 a,

AJUSNUN II..

C9Y 0

0-~p AlS3

271

II

333£333 0

34�

3 (0I

3U

'U

3 .014�3 0

a,* S.-

0� 0

0.

8I�

28

:-

1

I

04Sm

www -

4J -

0)J

29-

w an

01 0

-0020 0.16 a

* I0099 00

)4J> eU

WEIn

0s 1 1 0 E r

CL 00 4.

r-

010

091 '4-0

o~l 0 NWi-r 91.--

30*

>'

0 cm

I!F I~II "1.0

.81

F,

.6

.4

.2

10 20 30 40 50 SO 70 sO 90 I1.0 110 1 200, CPM

Figure A-9. Fourier power spectral density plot of horizontal intensityprofile (afternoon propagation)

31

I€

I

S" O 9

0

0 _

CL 0

14) )=

-j --

0.0•

4-1inCL

LL.

40~ 0nin f

324

I4J

uU

I4A5

0!

CLW°4.1[U

41

P..

4JE>

Sac

4J

411U

40..

ARl

33.O

II

ImI4o

00

40

14-

u to

r_ C.

4 00

L.e

tot

NCL

04'

340

4J

Ito

ot

0 uC

0~

to 0.0

* '4JCL 0L.C

44,

351'

I

I

Figure A-14. Sample of beam cross-section data used for quantitativeanalysis (night propagation)

36

I

I

.9

.2

S.4.

.4

k .2

0 10 20 30 40 s0 60 70 s0 90

f, CPM

Figire A-15. Fourier power spectral density plot of horizontal intensityprofile (night propagation)

VJ

c£ 4

on

-41

0L.

LA-

on on l

36)

[g

II

40

CL

I-cii•

I>

060

F 0 4j

4-b

T4

a.q

Ch

39

CD1Si0a

N 1

Eul

rC

>o

u. 0

cC

W I

s-m

CQ

w,m0 Estn 0

4o-

4JII IC

Im C

M to

4J 90t

K CL 0.64.

4- IV

M*CL

~44J

utA I

o ucZ'nn

C: I*0 c'.- 41

'aL ch

UEU

00

a -4- U -

4- :4fV S- 4

FTI

iI

/1' 1*N 'U

4.)C0

* N*1�LC

CN 0C

CLa,

4.)

S

U

* C* 0

4.)

V5-

43

'V

E Q

a cJ- a

V *Oi

La.�0

0�

43

II

C.-* dO d

5-

444

99.99 -

a99.9 -

99.8 -

99 0

98 -0

95

90

s0

70

60

I.- 50zw 40

w 30

20 0

10

5

2

0.5

0.2 0

0.1

0.0

0 .0 1I I , I I I i i i-1.0 -. 9 -. 9 . -. 6 -. 5 -. 4 -. 3 -. 2 -. 1 0 .1 .2 .3 .4 .5 .6 .7

LOG I

Figure A-23. Cumulative distribution curve for afternoon horizontalintensity scan 45

i -

lI!I

9.94

99.l

99

95

so"£ 70

S60

50Zlw 40

w1 30I.

20

0I0

0

5

2

00.5

0.2

0.10.05 -0.20-

•. 1 I I I i i I I i I I I I I i i I * j-I.0 -. 9 -. 8 -.7 -. 6 -. 5 -. 4 -. 3 -. 2 -. I 0 .I .2 .3 .4 .5 .6 .7

LOG IN -

Figure A-24. Cumulative distribution curve for night horizontalintensity scan 46

I]

S -1

UnclassifiedSecu•ty Classification

DOCUMENT CONTROL DATA. R&D

1.ORIGINATING ACTIVITY (CONFin0b AWOf) &I% PEPOR? 06CURITY CL*UPICATIOPU.S. Army Ballistic Research Laboratories Unclassified

Acerdeen Proving Ground, Maryland As. 60040P3 EPORT TI TLE'

OPTICAL METHOD FOR ANALYSIS OF ATMOSPHERIC EFFECTS ON LASER BEAMS

s. OgeCNIPTIvE "OTC* J1ppe *I.lanAhosa•.l dlv•fitl)

S. Atj THORIS) (PS awl ANW. 014M DarbS, 5a1 ASO&e)

Paul H. Deitz

a. 14P1ONT DATE I& TOTAL 0#0. OF PASKA 7b. NO. OFl Napo

, July 1967 60 3S.. CONTNACT Oa GNANT NO. I&. ORIGINATOWS NEPORT N1, WNWI001

b.• ,NOJI.o.-€,, RDT&E IP523801A286 Memorandum Report No. 1840

C. u. OT14M EPOR MuOMnw~ (AIW 04M ainbb- 16# a be .end-

1O. 9Y•IOUTIO" STATCUaNT

Distribution of this document is unlimited.

sented at the Symposium on Modern Optics,'

* sponsored by Brooklyn Polytechnic Institute U.S. Army Materiel Commandand he~d at Waldorf-Astoria Hotel, New York Washington, D.C.Cit.2l1-g4 March 1967.--Si•~~1. A19PTNACT • •mmmm

Ibis- report describes the design of an instrumentation system for the study of thef effects of atmospheric turbulence on a collimated laser beam under near-earth

conditions.

The instrumentation consists of a helium-neon laser with optical collimator and areceiving system of 24-inch aperture with narrow bandpass filter.

The Appendix presents axamples of beam cross section patterns for different propaga-tion corditions. The method of analyzing spatial intensity distributions of thesepatterns is described.

~~ftin.Aw OOMLaE. e ans.am*SG IOTA 4 73 UnclassiL'ied

. .-

Laser PropagationAtmospheric TurbulenceMissile GuidaniceComunications

U ca

I "

LaserUnroassitien

-~ 14

I T